ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/Coro/Changes
Revision: 1.270
Committed: Tue Oct 9 02:15:58 2007 UTC (16 years, 7 months ago) by root
Branch: MAIN
Changes since 1.269: +2 -0 lines
Log Message:
Coro::Util::inet_aton did not short-circuit dotted quad forms, causing a fork per resolve

File Contents

# Content
1 Revision history for Perl extension Coro.
2
3 TODO: should explore PerlIO::via::CoroCede and PerlIO::via::CoroAIO.
4 TODO: maybe implement a default message channel, very much like Erlang
5 (which is cool in a lot of important aspects (failures!),
6 but very lacking in others (higher level ipc)).
7
8 4.1
9 - incompatible change: $SIG{__DIE__} and $SIG{__WARN__} will now
10 be local to each coro (see Coro::State).
11 - possibly bring back 5.10 compatibility (untested).
12 - work around stupid (and wrong) warning on 5.10 :(.
13 - overlay the saved state over the context stack. This saves
14 a few hundred bytes per coroutine on average and also
15 speeds up context switching a bit.
16 - further tune default stack sizes.
17 - (more) correctly calculate stack usage in coro_rss.
18 - Coro::Storable::blocking_* did not properly lock
19 resulting in races between coroutines.
20 - added Coro::Storable::guard.
21 - stopping to trace a coroutine could destroy the cctx of
22 an unrelated coroutine.
23 - explain the relationship between Perl and C coroutines in
24 more detail in Coro::State.
25 - Coro::Util::inet_aton did not short-circuit dotted quad forms,
26 causing a fork per resolve. This also affected Coro::Socket.
27
28 4.03 Sat Oct 6 21:24:00 CEST 2007
29 - added Coro::throw method.
30 - minor code cleanups.
31
32 4.02 Sat Oct 6 02:36:47 CEST 2007
33 - fix a very minor per-coroutine memleak (a single codereference).
34 - fixed a bug where the currently in-use c context would be freed
35 prematurely (can happen only when programs change the stacksize
36 or use tracing).
37 - tracing can no longer keep a coro alive after it terminated.
38 - do static branch prediction in the common path for gcc. gives
39 about 2-5% speed improvement here.
40
41 4.01 Fri Oct 5 22:10:49 CEST 2007
42 - instead of recreating *a* standard output handle we simply
43 use STDOUT, which is faster and hopefully more robust.
44
45 4.0 Fri Oct 5 12:56:00 CEST 2007
46 - incompatibly changed Coro::Storable::freeze.
47 - major new feature: added Coro::Debug, for interactive coroutine
48 debugging, tracing and much more.
49 - major bug fix: unbelievable, but true: $_, $/ and many other
50 "saved" variables actually weren't being saved. This has been fixed,
51 of course, while increasing performance while losing all the save
52 flags.
53 - save flags are gone, and all the api functions dealing with them.
54 - added Coro::Semaphore::adjust.
55 - added Coro::Util::fork_eval.
56 - added Coro::Storable::{nfreeze,blocking_{freeze,nfreeze}}.
57 - added Coro::killall.
58 - reduce initial stack sizes to allow for "micro-coroutines".
59 - better async_pool resource management, moved parts of async_pool
60 handling to XS (major speed improvement).
61 - actually croak before modifying important data structures.
62 - refuse to transfer while compiling.
63 - possibly support eval EXPR better now.
64 - enable assembly per default on linux+bsd x86+amd64.
65 - all internal members were renamed _something for easier subclassing.
66 - many minor tweaks.
67
68 3.63 Wed May 16 14:10:06 CEST 2007
69 - implement handcoded assembly for x86/amd64 SVR ABI.
70
71 3.62 Fri Apr 27 21:36:06 CEST 2007
72 - upgrade libcoro (which might set unwind info correctly).
73 - change default on linux to setjmp/longjmp.
74
75 3.61 Thu Apr 19 12:36:18 CEST 2007
76 - Coro::Storable caused an endless loop when thawing invalid
77 pst's sometimes.
78 - use a Semaphore in Coro::Storable, as Storable doesn't
79 seem to be reentrant (although it is documented to
80 be threadsafe...).
81 - fix Coro::Signal to bring back the original unreliable
82 but stateful semantics.
83 - fixed a lot of typos in Coro.pm (patch submitted by David
84 Steinbrunner, which applied flawlessly).
85
86 3.6 Sat Apr 14 17:13:31 CEST 2007
87 - added some bugfixes to get eg/myhttpd working again.
88 - added Coro::Storable for often-cede'ing freeze/thaw.
89 - try to do a clean exit when a coroutine calls exit
90 (EXPERIMENTAL).
91 - got rid of indirect call through _coro_init.
92 - updated the partly antique examples in eg/ to
93 work again and be a bit less magic, too.
94 - fixed Coro::Signal semantics to work as documented again.
95
96 3.55 Sun Mar 25 01:20:47 CET 2007
97 - add SAVE_DEFFH to save the default filehandle and enable
98 it by default.
99 - finally move socket-operations from Coro::Socket to Coro::Handle
100 to be able to unblock foreign sockets.
101 - add Coro::State::save_also and guarded_save.
102 - add count accessor to Coro::Semaphore.
103 - add Coro::State::cctx_stacksize.
104 - just for the fun of it, do not rely on implicit context,
105 which can dramatically improve performance, but people
106 using windows-process-emulation perls do not care much about
107 performance.
108
109 3.51 Sun Mar 4 14:18:04 CET 2007
110 - fixed a problem when you weakened references to Coro::State's
111 (patch by Yuval Kogman).
112
113 3.501 Wed Feb 28 12:44:07 CET 2007
114 - rename some global symbols as macosx from hell redefines
115 them without asking.
116
117 3.5 Tue Feb 13 20:22:53 CET 2007
118 - do AnyEvent model detection earlier, avoiding problems
119 caused by first using AnyEvent and later Coro::Event.
120 - implement and document Coro::Event event objects.
121 - fix a potential problem in Coro::Event causing crashes.
122 - initialise PL_comppad when creating a new coroutine,
123 avoids crashes on early coro destruction.
124
125 3.41 Mon Jan 22 19:19:49 CET 2007
126 - readline on Coro::Handle did not support undefined $/,
127 nor did it deliver partial lines on EOF or error.
128 - implement malloc fallback for stack allocation because
129 stupid broken idiotic OSX has a stupid broken
130 idiotic fits-the-whole-os mmap "implementation" and
131 my dick feels longer if Coro is portable even to
132 obsolete platforms.
133
134 3.4 Fri Jan 19 21:52:54 CET 2007
135 - remove t/09_timer.t, as it isn't really testing much
136 but was rather flaky in practise.
137 - async_pool coro would keep arguments and callback alive until
138 it was reused.
139 - cancellation of a coroutine could cause spurious idle calls
140 in cede_notself.
141
142 3.3 Sat Jan 6 03:45:00 CET 2007
143 - implement $coro->on_destroy.
144 - Coro::Event blocking semantics have been changed,
145 documented and - hopefully - improved.
146 - fix nice adding, not subtracting, from priority.
147 - fix ->prio and api_is_ready (patch by Mark Hinds).
148 - fixed an assert ("... == prev__cctx->idle_te")
149 that could errronously trigger.
150 - fix various large and small memleaks.
151 - use a (hopefully) more stable cancel implementation
152 that immediately frees the coroutine data.
153 - cede/cede_notself return a status now.
154 - added Coro::guard function.
155 - added a global coroutine pool for jobs (on my machine,
156 I can create and execute 48k simple coros/s with async,
157 and 128k coros with async_pool).
158 - Coro::AIO now uses the coroutine priority as io priority.
159
160 3.2 Fri Dec 22 05:07:09 CET 2006
161 - improve portability to slightly older perls.
162 - use cleaner coroutine destruction.
163 - simplify configuration for users.
164 - optionally (unrecommended) prefer perl functions over
165 their coro replacements.
166
167 3.11 Tue Dec 5 13:11:24 CET 2006
168 - fixed some bogus assert's, but as perl.h disables assert even
169 without NDEBUG (thank you very much), not too many people should
170 notice (that did include myself). Andreas König noticed, though :)
171 - do not save/restore PL_sortcxix on >= 5.9.x, it doesn't seem to have
172 it. Also noticed by Andreas König :)
173 - save/restore tainted status.
174 - verified to pass the testsuite on my 5.9.5.
175
176 3.1 Mon Dec 4 23:03:40 CET 2006
177 - INCOMPATIBLE CHANGE: $/ is now per-coroutine (but slow).
178 - incompatible change: transfer flags are now per-state.
179 - give a better error message on deadlock.
180 - document Coro::nready.
181 - enhanced testsuite.
182
183 3.01 Sun Dec 3 23:47:42 CET 2006
184 - forgot to include Coro::Timer.
185
186 3.0 Sun Dec 3 22:57:25 CET 2006
187 - the "FINALLY COMPLETELY STABLE" release (coming soon:
188 the "FAMOUS LAST WORDS" release).
189 - implement a new stack sharing algorithm, which uses a stack
190 pool (size currently hardcoded).
191 - make stack sharing mandatory (it no longer uses a heuristic).
192 - eval/die no longer cause weird problems under heavy use.
193 - Coro::Event could cause livelocks if it was used but
194 no Coro::Event watchers were used.
195 - Coro::Event now uses asynccheck as prepare does not
196 check for changed watchers.
197 - Coro::Event allows multiple waiting coros and will wake up one
198 per event.
199 - Coro::Event should be cleaner and more efficient now.
200 - new utility function Coro::unblock_sub.
201 - document the sad fact that Event is no longer reentrant.
202 - putting a coroutine into the ready queue twice could under
203 some circumstances lead to stack corruption.
204 - minor incompatibility: subclassing Coro::State is supported
205 directly now without going through a _coro_state member.
206 - state/coro switching is much faster now.
207 - very minor optimisations and code/documentation cleanup.
208 - avoid problems due to compiler inlining.
209 - removed timers from Coro::Timer -> use AnyEvent instead.
210 - replaced Coro::idle coroutine by (cleaner) idle handler.
211 - updated to newest libcoro.
212 - implement enhanced support for valgrind.
213 - implement is_ready and return value for ready.
214 - removed Coro::Cont, it was a misnomer (it's generators),
215 and it was somewhat annoying to get it right. Will come back
216 if somebody asks for it :)
217 - many ->wait methods and Coro::Event could return
218 spuriously without the event having happened.
219
220 2.5 Tue Nov 7 12:22:33 CET 2006
221 - made Coro::Util, Coro::Select, Coro::Handle and Coro::Socket
222 use AnyEvent, moved them to Coro/.
223 - added Coro::LWP which contains all the uglyness required to
224 make LWP non-blocking.
225 - should work with perl 5.9.x now (Andreas König made me do it).
226 - fixed another bug in Coro::Select when the timeout was undef.
227 - reuse PL_start_env for all coros, saves some memory per coroutine.
228 - manage PL_top_env differently, hopefully to avoid panic: top_env.
229 - timeout argument was not properly used in Coro::Socket.
230 - allow limited forms of subclassing in Coro::Handle/Coro::Socket.
231 - emulate undocumented(!) functionality of IO::Socket required
232 by LWP(!!).
233 - updated eg/lwp to work with newer lwp's.
234 - remove "FATAL: uncaught exception" prefix. Coroutines that die
235 kill the whole process, just as exceptions in the main "coroutine"
236 did already.
237
238 2.1 Wed Nov 1 23:01:13 CET 2006
239 - fix a long-standing bug in Coro::Select where select with
240 zero timeout would instead change the current default filehandle.
241 - use a simpler and hopefully more robust way to clone padlists
242 (uses less memory and a perl function instead of our own).
243 - coro can now create a stack guard area on many architectures.
244 - Coro::AIO properly reexports additional functions from IO::AIO.
245 - updated libcoro with a workaround for OS X,
246 pach and testing by Michael Schwern.
247
248 2.0 Tue Oct 24 05:47:17 CEST 2006
249 - support additional aio requests in Coro::AIO.
250
251 1.9 (never properly released due to a glitch)
252
253 1.8 Thu Feb 2 00:59:06 CET 2006
254 - applied suggested patch by SAMV to avoid problems during stupid
255 mark & sweep gc run.
256 - applied patch by Scott Walters for 5.9.3 compatibility.
257
258 1.7 Tue Dec 27 01:41:58 CET 2005
259 - added Coro::AIO, a thin wrapper around IO::AIO.
260 - improved Makefile.PL explanations.
261
262 1.6 Mon Dec 12 21:30:05 CET 2005
263 - additionally save PM_curpm might fix as-of-yet
264 unseen problems with regex matches being attributed to the
265 wrong package.
266 - add t/10_bugs.t, which currently checks against imho broken
267 perls that use 0.26MB of stack space per Perl_magic_get
268 invocation (newer linuxes) as opposed to the 0.0002MB perl
269 normally uses.
270 - make stacksize configurable for the ultimate debian experience.
271
272 1.51 Mon Dec 12 18:48:36 CET 2005
273 - remove debugging warn() accidentally left in Coro::Select.
274
275 1.5 Tue Nov 29 13:32:44 CET 2005
276 - use Coro::Event inside Coro::Select to avoid spurious deadlocks.
277 - fix Coro::Select 'select'.
278 - strict'ify some modules.
279 - libcoro errornously restored the SIGUSR2 handler to SIGUSR1.
280 - use XSLoader in selected modules.
281 - remove some 5.6 compatibility cruft.
282
283 1.4 Tue Sep 6 00:11:05 CEST 2005
284 - libcoro did not take into account the trampoline on amd64,
285 when 'l'inux method was used.
286
287 1.31 Tue Aug 30 23:31:33 CEST 2005
288 - some portability fixes/workarounds.
289
290 1.3 Sat Aug 20 03:08:56 CEST 2005
291 - no code changes. module seems to work fine.
292 - improve Coro::State docs, remove reference to nonexisting
293 Coro::State::flush method.
294 - no longer autodetect windows, present it as an option instead.
295
296 1.2 Mon May 16 02:00:55 CEST 2005
297 - included libcoro.c earlier in State.xs, might improve portability.
298 - use a faster and possibly more stable Coro::Cont implementation.
299 - accept x86_64 in addition to amd64 for optimized linux-amd64 support.
300 - fix bugs Coro::SemaphoreSet that could cause locks never to be freed.
301 - fix bugs in CoroAPI.h and document it in Coro::MakeMaker.
302
303 1.11 Thu Mar 3 18:00:52 CET 2005
304 - change of contact address.
305
306 1.1 Tue Feb 22 20:51:16 CET 2005
307 - support [l]inux method on amd64.
308 - allow some context switches while compiling/eval'ing.
309
310 1.0 Tue Aug 10 03:47:27 CEST 2004
311 - Event 0.89 fixes the bug again.
312 - fix the remaining(??) memleak. I didn't free the cache-AV for
313 padlists, so one AV per created anon-sub leaked.
314
315 0.97 Fri May 14 15:23:32 CEST 2004
316 - remove stupid gcc extension that was an experiment only anyway.
317
318 0.96 Thu May 13 18:09:29 CEST 2004
319 - tell people if they have yet another broken Event variant.
320 - made Coro to work _better_ with threads (it seems that, as long
321 as you use Coros in one thread only everything will work fine.
322 Looking at the perl source the problem might be missing locking
323 between Coro and some perl internal routines. That's not easily
324 fixable).
325 - (hopefully) got rid of the static xnext inside transfer.
326
327 0.95 Sun Feb 15 03:20:28 CET 2004
328 - removed Coro::State::flush, implemented a new and much
329 faster caching mechanism by attaching magic to the CV.
330 - a memleak with (real) closures remains.
331 - dramatically reduced size of stacks to 65536 (linux/x86), after
332 observing that even myhttpd never had more than a kilobyte
333 of stack in use (stack is still only physically allocated in
334 pagesize increments for systems suporting mmap).
335 - fix for a "cygwin-64" architecture added, proposed by Stefan Traby.
336 - fix for threaded perl.
337
338 0.9 Sun Nov 23 04:32:39 CET 2003
339 - fix a few very hard to track down but (of course) stupid
340 memory leaks.
341 - stringify version number to avoid locale problems :(.
342 - perl 5.9.0 does not have SvPADBUSY, reported by Scott Walters,
343 hopefully ignoring it "just works".
344
345 0.8 Wed Nov 5 19:38:40 CET 2003
346 - port to perl5.005_03, but only 5.8.x+ is supported!
347 - honor LocalAddr even without LocalPort in Coro::Socket.
348
349 0.7 Tue May 27 03:12:38 CEST 2003
350 - the version jump indicates some level of testing, not gobs
351 of new features.
352 - uh... I found the prompt function in ExtUtils::MakeMaker.
353 Highly correct stuff, that is...
354 - fixed(?) a bug with die's in coroutines causing "panic: top_env".
355 the fix is not well understood by the author, so beware :(.
356
357 0.652 Thu May 8 02:54:46 CEST 2003
358 - Applied patch by Slaven Rezic to set default to "s" on FreeBSD,
359 cause version 4 doesn't have ucontext.
360 - Benjamin Reed reported that setjmp works fine on darwin,
361 so preselect it.
362
363 0.651 Sat Mar 29 15:00:23 CET 2003
364 - fix a bug in Coro::Handle where some bytes could get lost
365 on reading, reported by jason@nichego.net.
366
367 0.65 Sun Mar 23 00:08:26 CET 2003
368 - added a README.linux-glibc.
369 - new module Coro::Select.
370 - also save/restore PL_comppad, fixes segfaults with 5.8.1.
371 (I never claimed that I know what I am doing ;).
372 - set default to setjmp/longjmp on non-x86-linux.
373
374 0.6 Thu Nov 21 11:09:06 CET 2002
375 - made lazy context switching the default.
376 - fixed the bug where SemaphoreSet::guard
377 was the same as timed_guard and timed_guard was missing.
378 - fix a memleak in Coro::Timer.
379 - reclassified context sharing as not experimental.
380
381 0.534 Sun Apr 14 03:05:12 CEST 2002
382 - fixed backspace => tab in header parsing (myhttpd).
383 - added eg/readline, for elmex.
384 - Coro::Event now calls ->start on first call instead of
385 ->again, which created wrong timeouts for the first ->next.
386 - fixed a bug where Coro::Socket returned a closed socket instead
387 of nothing on ECONNREFUSED and similar error conditions.
388
389 0.533 Mon Feb 18 18:49:40 CET 2002
390 - adapt to finally working Event-0.85.
391 - fixed buggy event initialization (trapped under windows, but how
392 could it possibly have worked anywhere?)
393
394 0.532 Wed Jan 16 02:45:32 CET 2002
395 - removed the reference to pp_entersub, might work on
396 (some) win32 perls now (testsuite works under cygwin,
397 without Event, which freezes).
398
399 0.531 Mon Dec 10 22:18:44 CET 2001
400 - Fixed a bug in SemaphoreSet::timed_down.
401
402 0.53 Tue Nov 27 21:11:13 CET 2001
403 - some tests for Event.
404 - slightly different internal architecture to get refcnt into the object
405 and not the reference. might fix a few bugs, certainly did introduce
406 new ones.
407 - Coro::Timer (independent of Event).
408 - new timed_wait functions for Coro::Signal, Semaphore, SemaphoreSet.
409
410 0.52 Tue Nov 6 21:36:18 CET 2001
411 - ported to cygwin (trivial).
412 patch by Gerrit P. Haase <gp@familiehaase.de>.
413 - small setjmp code fix by Sullivan.DanielJ@epamail.epa.gov.
414
415 0.51 Thu Nov 1 20:39:01 CET 2001
416 - terminate/cancel now work properly (otherwise termination
417 could cause a "next coroutine is not and contains not..."
418 error.
419 - added Coro::Socket::shutdown.
420 - Coro::Event::loop no is the same as Event::loop.
421 - implemented terminate with args + join.
422
423 0.5 Fri Sep 28 16:15:35 CEST 2001
424 - fixed "print" on a Coro::Handle. The print method worked.
425 - small tweaks (seem to reduce memory consumption a lot)
426 in various modules.
427 - splendid use of "no warnings" scattered throughout.
428 - added Coro::Handle::rbuf, fixed Coro::Handle::fh.
429
430 0.49 Sun Sep 16 02:42:45 CEST 2001
431 - changed some method calls to function calls for speed
432 inside Coro::Handle.
433 - make Coro::Handle use an array instead of a hash for
434 speed reasons.
435 - IRIX mystery solved: it's SGI's NT, after all: "standard, huh?".
436 sjlj and ucontext should now work.
437 - IRIX-specific port for libcoro.
438 - swapped order of accept results to match IO::Socket.
439 - changed getsock/peername to sock/peername to match IO::Socket.
440 - fixed a bug that caused segfault when returning to main task
441 under some circumstances.
442 - other bugfixes.
443
444 0.45 Sun Sep 2 02:54:01 CEST 2001
445 - new method Coro::Handle::timeout.
446 - corrected speling of set...name to get...name in Coro::Socket.
447 - Coro::Socket::accept now returns a Coro::Socket, not a Coro::Handle.
448 - Coro::Handle now supports fileno and FILENO.
449 - added eg/myhttpd, a web server.
450 - Coro::Socket now accepts numeric ports as well as "port(number)"
451 syntax.
452 - moved some scheduling primitives into xs code. more to come.
453 - new simple priority system.
454 - implemented Coro::Event using XS, almost four times faster!
455 - small memory corruption problem fixed (boy that was difficult).
456
457 0.13 Wed Aug 8 16:53:07 CEST 2001
458 - new method Coro::Handle::readline.
459 - added eg/lwp and Coro::Handle::autoflush, to make LWP work.
460 (see eg/lwp on how to make LWP non-blocking).
461 - renamed Coro::Socket::new_inet to new.
462 - added Coro::Util, some utility functions.
463 - die/eval now works better (still get annoying (but true ;)
464 "Callback called exit" messages).
465 - Coro::Handle now supports timeout, and Coro::Socket Timeout.
466 - much better Coro::Socket support.
467
468 0.12 Fri Jul 27 04:19:01 CEST 2001
469 - do not use mmap if neither MAP_ANON nor MAP_ANONYMOUS is defined.
470 - very experimental stack sharing algorithm. not 100% safe but
471 should work well in practise ;)
472 - added Coro::RWLock.
473 - Coro::Specific now works.
474
475 0.11 Tue Jul 24 22:49:21 CEST 2001
476 - added specialized hack for newer and older linux versions (fast).
477 - renamed Coro::Event::IO to Coro::Handle.
478 - new module Coro::Socket.
479
480 0.1 Tue Jul 24 01:47:53 CEST 2001
481 - release candidate 3. A rather new internal structure :(
482 - the great renaming: Damian Conway gave me a suitable replacement
483 for yield.
484 - added Coro/libcoro, a portable coroutine implementation for C
485 (not even perl-dependent!!!), which can be used by Coro::State.
486 - renamed SAVE_DEFSV (on xs-level only) to avoid symbol clash in
487 perl-5.6.
488 - new function Coro::Event::idle.
489 - the idle process is now overriden
490 by default in Coro::Event.
491 - Coro::Channel now enforces the size.
492 - canceling events now works in all cases.
493 - Coro state now include $_ and $@.
494 - yet another bug workaround that I do not really understand :(
495 - new module Coro::Event::IO (very undocumented) to do
496 non-blocking i/o.
497 - performance tuning.
498
499 0.08 Thu Jul 19 06:13:25 CEST 2001
500 - release candidate 2 (new functionality).
501 - add Coro::State::flush function.
502 - transfer now supports a flags value (mostly for speed ATM).
503 - might compile and work in the presence of threads now.
504 - continuations have a saner syntax.
505 - no more memleaks.
506
507 0.07 Tue Jul 17 17:40:18 CEST 2001
508 - release candidate 1 ;)
509 - slightly nicer code.
510 - fixed a scheduling bug in Coro::Event.
511
512 0.06 Tue Jul 17 04:23:24 CEST 2001
513 - ok, I found the showstopper - the same sub must not be
514 re-entered in two different coroutines, otherwise => crash. I
515 see no easy solution to this problem, except by walking the call
516 chaing and saving/restoring the cv's, which is what I do now.
517 - memory leaks still latent, especially at thread termination.
518 - Coro::Event now works (haha).
519
520 0.05 Sun Jul 15 17:32:20 CEST 2001
521 - fixed a few issues in Coro::Event.
522 - I forgot to include Coro::Event in 0.04 :(:(:(
523
524 0.04 Sun Jul 15 05:24:59 CEST 2001
525 - @_ is now properly localized.
526 - Coro::State is now easier subclassable.
527 - Coro::Cont now coroutine-aware.
528 - Coro::Specific is a low-overhead module to create
529 coroutine-specific vars.
530 - Coro::Event provides a simple interface to Event.
531
532 0.03 Fri Jul 13 14:51:52 CEST 2001
533 - transfer() now implemented in XS (beware).
534 - new module Coro::Cont for really faked continuations.
535 - big internal architecture changes: Coro::State is now
536 really low-level and can thus be used to implement other
537 interesting things, While "Coro::" implements a process-like
538 model. Still crude and subject to change.
539 - $_ and $@ are no longer being localized.
540
541 0.02 Tue Jul 10 01:38:17 CEST 2001
542 - implemented "async" attribute.
543 - $_ and $@ are now localized.
544 - added Coro::Channel.
545 - more testcases, still no docs.
546
547 0.01 Tue Jul 3 02:18:41 CEST 2001
548 - original version; copied from Convert::Scalar.
549